Stratified Linkage Disequilibrium Score Regression reveals enrichment of eQTL effects on complex traits is not tissue specific
Authors:
- Hill F. IpHIHill F. Ip
- Rick JansenRJRick Jansen
- Abdel AbdellaouiAAAbdel Abdellaoui
- Meike BartelsMeike Bartels
- UK Brain Expression ConsortiumUCUK Brain Expression Consortium
- Dorret I. BoomsmaDBDorret I. Boomsma
- Michel G. NivardMNMichel G. Nivard
Preprint
Abstract
Both gene expression levels and eQTLs (expression quantitative trait loci) are partially tissue-specific, complicating the detection of eQTLs in tissues with limited sample availability, such as the brain. However, eQTL overlap between tissues might be non-trivial, allowing for inference of eQTL functioning in the brain via eQTLs measured in readily accessible tissues, e.g. whole blood. Using Stratified Linkage Disequilibrium Score Regression (SLDSR), we quantify the enrichment in GWAS signal of blood and brain eQTLs in genome-wide association study (GWAS) on 11 complex traits (schizophrenia, BMI, educational attainment, Crohn’s disease, rheumatoid arthritis, ulcerative colitis, age at menarche, coronary artery disease, height, LDL levels, and smoking behavior). Our analyses established significant enrichment of blood and brain eQTLs in their effects across all traits. As we do not know the true number of causal eQTLs, it is difficult to determine the precise magnitude of enrichment. We found no evidence for tissue-specific enrichment in GWAS signal for either eQTLs uniquely seen in the brain or whole blood. To follow up on our findings, we tested tissue-specific enrichment of eQTLs discovered in 44 tissues by the Genotype-Tissue Expression (GTEx) consortium, and, again, found no tissue-specific eQTL effects. We further integrated the GTEx eQTLs with SNPs associated with tissue-specific histone modifiers, and interrogate its effect on rheumatoid arthritis and schizophrenia. We observed substantially enriched effects on schizophrenia, though again not tissue-specific. Finally, we extracted eQTLs in tissue-specific differentially expressed genes, and determined their effects on rheumatoid arthritis and schizophrenia. We conclude that, while eQTLs are strongly enriched in GWAS signal, the enrichment is not specific to the tissue used in eQTL discovery. Therefore, working with relatively accessible tissues, such as whole blood, as proxy for eQTL discovery is sensible; and restricting lookups for GWAS hits to a specific tissue might not be advisable.Keywords
All Related Versions
- Published version: Behavior Genetics, 48 (5), 374.
This publication has 47 references indexed in Scilit:
- Local genetic effects on gene expression across 44 human tissuesbioRxiv, 2016
- The molecular hallmarks of epigenetic controlNature Reviews Genetics, 2016
- LD Score regression distinguishes confounding from polygenicity in genome-wide association studiesNature Genetics, 2015
- Genetic pleiotropy between multiple sclerosis and schizophrenia but not bipolar disorder: differential involvement of immune-related gene lociMolecular Psychiatry, 2014
- Partitioning the Heritability of Tourette Syndrome and Obsessive Compulsive Disorder Reveals Differences in Genetic ArchitecturePLOS Genetics, 2013
- Gene Expression in Skin and Lymphoblastoid Cells: Refined Statistical Method Reveals Extensive Overlap in cis-eQTL SignalsThe American Journal of Human Genetics, 2010
- Association analyses of 249,796 individuals reveal 18 new loci associated with body mass indexNature Genetics, 2010
- Hundreds of variants clustered in genomic loci and biological pathways affect human heightNature, 2010
- Genome-wide association of major depression: description of samples for the GAIN Major Depressive Disorder Study: NTR and NESDA biobank projectsEuropean Journal of Human Genetics, 2008
- Genetic Dissection of Transcriptional Regulation in Budding YeastScience, 2002